Usability of 'samba-tool domain provision'
repenny241155 at gmail.com
Thu Jun 2 07:56:54 UTC 2016
On 02/06/16 08:32, Andrew Bartlett wrote:
> On Thu, 2016-06-02 at 09:24 +0200, Andreas Schneider wrote:
>> On Thursday, 19 May 2016 14:25:21 CEST Rowland Penny wrote:
>>> OK, I given up with the 'spaghetti knitting' for the moment, the
>>> spaghetti is winning and I needed a rest, so I had a look at this :
>>> I have split the patch into several smaller patches:
>>> The first patch makes the required changes to only provision a DC,
>>> is because, as far as I an aware, this is the only thing you can
>>> The second patch makes 'realm' & 'domain' required arguments, they
>>> no longer options.
>>> The third patch checks if the NetBIOS domain name is too long and
>>> if it is. This also fixes bug 11925
>>> The fourth patch adds Administrator password checking
>>> The fifth patch adds a couple of extra questions to 'interactive',
>>> use_xattrs and use_rfc2307
>>> The sixth and final patch adds a few examples to help.
>> I've reviewed the patches and did some small fixes. I've also fixed
>> so we can provision an AD DC.
>> However syntax of 'samba-tool domain provision' changes. I'm not sure
>> others are OK with this.
> We can't change the syntax. This is already in use in many scripts and
> automated tools. That the selftest needed to be changed is a red flag
> for me.
The whole thing that started this was that the NetBIOS domain name
didn't have enough 'help', the same went for the realm name, and this
cannot really be fixed without a total re-write of 'samba-tool'. This is
because it comes from 'sambaopts'
Why can't we change the syntax? just because external scripts use
samba-tool as it is now, isn't really a good enough reason, do we
totally guarantee that samba-tool will never change?
I seem to remember a similar argument about the 'debug' library and this
More information about the samba-technical